The unique pattern of class I major histocompatibility complex (MHC) antigen expression at the maternal/fetal interface is thought to play a major role in protecting the fetus from rejection by the mother's immune system. Trophoblasts, the only fetal tissue that comes into direct contact with the mother, do not express polymorphic class I and II MHC antigens. This may in part explain the lack of maternal immune response against the fetus. Recently it was found that a non-classical MHC molecule, designated HLA-G, is present on certain subpopulations of trophoblasts, suggesting that the expression of this non-polymorphic MHC molecule plays an active role in maternal/fetal tolerance. The transcript encoding HLA-G has been shown to undergo alternative splicing to give rise to a full length form as well as two smaller forms, all of which contain the alpha1 alpha helix. In addition, some of these forms have been shown to be secreted by cells. These observations, as well as studies from this laboratory using synthetic peptides corresponding to specific regions of classical MHC class I molecules, constitute the experimental basis for the hypothesis that soluble HLA-G peptides contribute to maternal immune unresponsiveness towards the fetus. A synthetic peptide corresponding to the entire alpha1 alpha helix (residues 60-84) of HLA-G inhibits T cell proliferation in vitro and also blocks IL-2 production by T cells. The aims of this proposal are to investigate the molecular mechanisms by which soluble HLA- G peptides induce unresponsiveness and to investigate their role in pregnancy. Specifically, studies are proposed: 1) to characterize the pathways by which synthetic peptides corresponding to HLA-G transduce a negative signal in human lymphocytes; 2) to identify the receptor to which these peptides bind; and 3) to examine the effects of these peptides in a rodent model of spontaneous abortion.